serieslisted

Fetch a Bank of Korea economic statistic as a time series from ECOS. Holds no logic of its own -- it calls the pyecos package's CLI (`ecos series`) and shows the result to the user. Needs a stat table code and usually an item code (find them with the catalog skill). Trigger phrases: 기준금리 시계열, 환율 추이, ECOS 통계 가져와, economic statistic time series, Bank of Korea interest rate, CPI series, 한은 통계.
seokhoonj/pyecos · ★ 0 · Data & Documents · score 70

Install: claude install-skill seokhoonj/pyecos

# ecos — statistic time series Take an ECOS statistic table code (and usually an item code) and print its observations over time. The fetching, pagination, and parsing live in the pyecos package (on PyPI); this skill is a thin wrapper that calls its CLI and relays the result. A rejected key or a vendor error comes back from the CLI as a one-line `ecos: <message>` -- relay it as-is rather than throwing a stack trace at the user. (A missing or malformed argument is caught earlier by argparse, which prints usage and exits 2.) ## Prerequisite This plugin calls the `ecos` CLI, so the package must be installed and an API key set: ``` pipx install pyecos # or: pip install pyecos export ECOS_API_KEY=... # a free key from https://ecos.bok.or.kr/api/#/ ``` That puts the `ecos` command on PATH. The key can also be stored in `~/.config/pyecos/credentials.json` as `{"ECOS_API_KEY": "..."}`. Without a key the CLI exits with `ecos: no ECOS API key ...`; relay that and point the user at the ECOS site. ## Running ``` ecos series "<STAT_CODE>" [--item CODE] [--cycle ...] [--start ...] [--end ...] [options] ``` Options (`ecos series --help` is the source of truth): - `--item CODE` — an item code selecting a sub-series; repeat up to 4 times. - `--cycle annual|semiannual|quarterly|monthly|semimonthly|daily` — frequency (default: monthly). - `--start` / `--end` — cycle-formatted period bounds: `202401` (monthly), `2024` (annual), `2024Q1` (quarterly), `20240115` (daily). E